Accelerate AI layout
In terms of artificial intelligence, Intel has developed a complete enterprise AI stack (Enterprise AI Gaudi 3 stack), using open source standards such as OpenVINO and Intel Developer Cloud oneAPI & Synapse to promote the rapid development of hardware, systems, and applications.
Hardware splits into: Data Center (scalable systems, accelerators, CPUs), Networking (open standards, infrastructure), and Client & Edge (AI PCs, NPUs, GPUs, CPUs).
Gaudi 2 is more cost-effective
Intel claims Gaudi 2 matches Nvidia’s A100 in price, yet surpasses it by 3x in performance, promising cost savings.Gaudi 2 hardware also outperforms the latest H100 GPUs in certain workloads, and Intel says it will continue to expand on these advantages to maintain its lead in the future.
Gaudi 3 is scheduled to be released in the third quarter of 2024
In terms of roadmap, Intel is still positioning Gaudi 3 as the follow-up to Gaudi 2 and Falcon Shores. Intel said the Gaudi 3 accelerator is currently in validation labs and will be generally available in the third quarter of 2024.
Intel said that in terms of software/applications, it has built an extensive OEM ecosystem and Intel Developer Cloud Network for Gaudi 3. In terms of hardware, the Gaudi 3 accelerator will use the same high-performance architecture as Gaudi 2, but has 4 times the computing power. , the network bandwidth is 2 times, and the HBM memory bandwidth is 1.5 times.
Falcon Shores plans to launch in 2025
Falcon Shores blends Gaudi AI IP with Intel GPU tech, pioneering an AI-focused GPU architecture. Development of Falcon Shores is progressing well, with a launch targeted for 2025.
Model migration can be achieved in 3-5 lines of code
AI developers worry about code modification when migrating to new hardware with different accelerators. Intel asserts smooth app migration by adding a few lines to Python scripts in higher-level stacks.
Intel’s “Model Migration” requires only 3-5 lines of code to migrate existing or new models to new hardware. For advanced software, Intel offers migration support from NVIDIA to the entire Gaudi and even next-generation Falcon Coast accelerator lineup.
Support for OpenAPI, Triton, and TPC-C will ease migration from NVIDIA hardware to Falcon Shores GPU.
Gaudi hardware tailored for the Chinese market set for release.
Intel said it will customize Gaudi hardware for the Chinese market and export it in full compliance with U.S. authorities.
